Biol Invasions (2019) 21:2871–2890

https://doi.org/10.1007/s10530-019-02019-4 (0123456789().,-volV)( 0123456789().,-volV)

ORIGINAL PAPER

Recruitment of native parasitic wasps to populations of the invasive in the northeastern United States

Hannah J. Broadley . Robert R. Kula . George H. Boettner . Jeremy C. Andersen . Brian P. Griffin . Joseph S. Elkinton

Received: 20 September 2018 / Accepted: 26 May 2019 / Published online: 7 June 2019 Ó Springer Nature Switzerland AG 2019

Abstract Ecological communities may be resistant to study the effect of recruitment of native parasitoids to invasive species through a combination of top-down on an invasive population of winter moth in the and bottom-up mechanisms, including predation, northeastern United States. We deployed sentinel competition, parasitism, and disease. In particular, pupae over 4 years across this population’s range, natural enemies that cross over from native species to identified recovered parasitoids, and measured the rate use newly introduced non-native species as hosts can of parasitism by native sources across years, seasons, influence invasive species population dynamics and invasion history, and host densities. Native Pimpla may slow down invasions. However, research on wasps inflicted 98% of the parasitism detected, parasitism in biological invasions is lagging behind resulting in an annual average of 15–40% mortality research on biological invasions in general. We used on pupae not depredated. Pimpla were present across the model species winter moth (Operophtera brumata) all years, seasons, and sites. Where winter moth has invaded, parasitism was greatest when winter moth pupal density was high (i.e., positive density-depen- Electronic supplementary material The online version of dent mortality) suggesting that Pimpla is helping to this article (https://doi.org/10.1007/s10530-019-02019-4) con- regulate the population. The wasps were morpholog- tains supplementary material, which is available to authorized ically identified as Pimpla aequalis Provancher; how- users. ever, using a multilocus genetic comparison approach, H. J. Broadley (&) Á J. S. Elkinton they were determined to comprise two cryptic species. Graduate Program in Organismic and Evolutionary Overall, this study shows that recruitment of these Biology, University of Massachusetts, Amherst, native wasps to the invasive winter moth population is MA 01003, USA likely playing a significant role in regulating popula- e-mail: [email protected] tion outbreaks and is aiding in biological control of R. R. Kula winter moth. Systematic Entomology Laboratory, Beltsville Agricultural Research Center, Agricultural Research Keywords Operophtera brumata Pimpla Service, U.S. Department of Agriculture, c/o National Á Á Museum of Natural History, Smithsonian Institution, Á Biotic resistance Á Parasitoid Á Washington, DC 20013-7012, USA Biological control

G. H. Boettner Á J. C. Andersen Á B. P. Griffin Á J. S. Elkinton Department of Environmental Conservation, University of Massachusetts, Amherst, MA 01003, USA 123 2872 H. J. Broadley et al.

Introduction Scotia and British Columbia using the tachinid fly Cyzenis albicans (Falle´n) (Diptera: Tachinidae) (Em- The introduction of non-native species to new eco- bree 1966; Murdoch et al. 1985; Roland and Embree logical communities is creating novel and altered 1995), similar efforts have been initiated in this most predator–prey and parasite–host interactions (Faillace recent invasion (Elkinton et al. 2015). While impor- et al. 2017; Garnas et al. 2016; Hobbs et al. 2009; tation biological control of winter moth in the Pearson et al. 2018; Shea and Chesson 2002; Strauss northeastern United States has shown promising et al. 2012). The species richness of a community can results (Elkinton et al. 2015, 2018), as previously predict the chance that an invasive species will found in Canada (Roland 1988, 1990), the overall successfully establish. This hypothesis, known as the success will likely depend on additional mortality biotic resistance hypothesis, holds that communities from native natural enemies. Parasitoid recruitment may resist invasions through a combination of factors from related native species can have significant effects including predation, competition, parasitism, and on invasive populations in other study systems disease (Elton 1958; Jeschke et al. 2012; Levine (e.g. Duan et al. 2013, 2014; Grabenweger et al. 2010; ` et al. 2004; Maron and Vila 2001; Sakai et al. 2001; Matosevic and Melika 2013; Schonrogge et al. 1995; Shea and Chesson 2002). Natural enemies of native or Zappala et al. 2012). This is especially true if the resident species that cross over to use non-native parasitoids limit the establishment or facilitate the species can influence invasions in complex ways eradication of a new exotic species (Tobin et al. 2011), (Dearborn et al. 2016; Faillace et al. 2017; Graben- if they respond to increases in the invasive popula- weger et al. 2010; Strauss et al. 2012) and have the tion’s densities with density-dependent mortality potential to slow down invasions and aid in biological (Holling 1973), or if the parasitoids parasitize at high control (Dearborn et al. 2016; Kenis et al. 2008; Maron rates at the invasion front such that they can slow or et al. 2001; Vindstad et al. 2013). These interactions stop the spread of the invasive population (Lewis and are particularly strong for non-native species with Kareiva 1993). sympatric native congeners and confamilials (Dear- Non-native species with native congeners in the born et al. 2016; Grabenweger et al. 2010; Strauss introduced range may be less likely to establish an et al. 2012; Vindstad et al. 2013). Further, biotic invasive population than species introduced to a range resistance may especially affect invasive without a native congener; these invasive species face because they are typically r-selected (Sakai et al. top-down pressure from the natural enemies of their 2001) and their population dynamics are closely congener (Callaway et al. 2013; Carrillo-Gavilan et al. related to natural enemies, predominantly parasitoids 2012; Diez et al. 2008; Keane and Crawley 2002; (Hassell 2000; Myers 2018; Waage and Greathead Richardson and Pysek 2006). The native congener of 1985). While it is clear that naturally-occurring winter moth, Bruce spanworm (Operophtera bruceata biological controls play an important role in pest Hulst), is a potential source of native parasitoid suppression (Heimpel and Mills 2017), research on the recruitment to invasive populations of winter moth role of parasitism in biological invasions is lagging in North America. Bruce spanworm is present in all behind research on biological invasions in general regions winter moth has invaded. In addition to having (Poulin 2017) and particularly so for studies of similar life-cycle dynamics, these two congeners use invasive herbivores and their natural enemies (Bu¨rgi similar hosts, are present at similar times of the year, et al. 2015). and can hybridize in the field (Gwiazdowski et al. The European winter moth, Operophtera brumata 2013; Havill et al. 2017). Thus, it is likely that native L. (Lepidoptera: Geometridae), is a well-known natural enemies that parasitize Bruce spanworm could defoliator of forest, shade, and fruit trees with multiple use winter moth as a host. Additionally, the life invasive populations established in North America histories of winter moth and Bruce spanworm make (Embree 1965; Myers 2018; Myers and Cory 2013; their populations particularly vulnerable to pupal Roland and Embree 1995), including a recent (circa mortality by predation and parasitism from native late 1990s) introduction in the northeastern United natural enemies. Both winter moth and Bruce span- States (Elkinton et al. 2010, 2015). Following suc- worm have a long pupation period (6–7 months during cessful biological control of winter moth in Nova the summer, representing the vast majority of its life 123 Invasive winter moth in the northeastern United States 2873 span) and pupates in the top layer of soil or leaf litter. association of Pimpla species with the invasive Together this renders both species particularly vul- population of winter moth in the northeastern U.S., nerable to pupal mortality by predation, parasitism, but neither study directly assessed Pimpla wasps as and disease. primary parasitoids of winter moth. The assemblage Parasitoid wasps in the genus Pimpla Fabricius and origin of Pimpla species associated with invasive (: Ichneumonidae) might be an impor- winter moth populations in the northeastern U.S. is tant source of mortality for invasive populations of unknown, as is their prevalence, role in causing winter winter moth in North America. Species of Pimpla are moth mortality, and potential to regulate winter moth found in all zoogeographic regions, including 27 densities. species in the Nearctic Region (Yu et al. 2012), and In this study we aimed to (1) quantify parasitism by typically parasitize Lepidoptera prepupae and pupae Pimpla wasps on winter moth pupae and C. albicans (Bennett 2008; Carlson 1979; Gauld 1991; Goulet and puparia across a spatial and temporal gradient, (2) test Huber 1993). Pimpla species are known to use for a density dependent effect of pupal parasitism, and geometrid pupae concealed in moss or soil (Fitton (3) identify the Pimpla species using morphological et al. 1988), suggesting that they may be important and molecular characteristics. We discuss our results natural enemies of Bruce spanworm and winter moth, in relation to their implications for understanding the which fit both of these criteria. Pimpla species are role and origin of this parasitoid in the control of an known from the region of this study and have been introduced lepidopteran pest. These findings demon- associated with winter moth populations. Pimpla strate the importance of evaluating native parasitoids turionellae L., Pimpla contemplator Mu¨ller, and in the establishment, spread, and population dynamics Pimpla flavicoxis Thompson have been recorded as of other invasive species. attacking winter moth in its native range in Europe (Fitton et al. 1988; Silvestri 1941; Wylie 1960). P. flavicoxis is currently considered a valid species (Yu Methods et al. 2012) but has also been treated as a junior synonym of Pimpla aquilonia Cresson (Carlson 1979). Pupal deployment P. aquilonia is a broadly distributed Holarctic species (covering the region of this study) reported from To acquire winter moth for deployment as sentinel multiple families of Lepidoptera, including geome- pupae, we collected winter moth larvae from long- trids, but not yet from winter moth (Yu et al. 2012). A term study sites across eastern Massachusetts each small release of P. turionellae to Nova Scotia, Canada spring from 2014 to 2017 (Elkinton et al. 2015). in 1955 was conducted to control winter moth; Larvae were reared in batches of 500 or fewer however, recovery collections revealed no evidence individuals in ventilated 20 L (5 gallon) buckets with of establishment (Graham 1958; Humble 1985). the foliage from the tree species on which they were Similarly, P. contemplator was released in Nova found. Mortality from viruses, other diseases, and Scotia, Canada in 1964 to control winter moth but with larval parasitism in these collections was minimal no evidence of establishment (Carlson 1979). Pimpla (Broadley et al. 2017; Donahue et al. 2018). When the hesperus Townes is native to North America and has larvae showed signs of pupating (thickening body been reported as a parasitoid of winter moth and Bruce shape), sifted peat moss was added to the bottom of the spanworm in British Columbia, Canada (Humble buckets for pupation. All winter moth pupae were non- 1985). Recently, an undetermined species of Pimpla destructively checked under a dissecting microscope in the United States in Maine was found to be more (Wild Heerbrugg M5 stereo) for parasitism by C. abundant at sites with high winter moth infestation albicans or other larval parasitoids. The unparasitized than moderately infested sites (Morin 2015), and winter moth pupae and pupae parasitized by C. another, or potentially the same, undetermined species albicans were stored at 12 °C in a growth chamber of Pimpla has been reported in the northeastern United (Percival Scientific) until deployment. States as a hyperparasitoid of C. albicans, an intro- To study pupal parasitism by native parasitoids, duced biological control agent of winter moth (Broad- winter moth pupae were deployed at sites across ley et al. 2018). These latter two studies revealed an eastern Massachusetts, Rhode Island, and Connecticut 123 2874 H. J. Broadley et al. from 2014 to 2017 (Table 1). The study sites were further parasitoid development. Pupae were stored at chosen to coincide with winter moth long-term study 12 °C until the beginning of December, at 9.5 °C until sites and to reflect a range of winter moth and C. the end of December, and at 4 °C until late March. The albicans establishment histories (Elkinton et al. pupae were kept in dark with no day/night cycle, and 2014, 2015). The study sites were all in mixed once a month they were sprayed with a sodium hardwood forests dominated by red oak (Quercus propionate solution (5 g sodium propionate/l of rubra). The pupae were deployed in three to five water) to prevent mold. Starting in late March, the consecutive rounds from mid-June until the end of temperatures were increased 4 °C every 3 days to October with five deployments (one every 3 weeks) in 22 °C when the pupae were taken out of storage and 2014 and three deployments (one every 6 weeks) in kept at room temperature. The parasitoids were 2015–2017. Each deployment consisted of 100 winter identified as Pimpla using keys in Townes et al. moth pupae attached to small burlap squares with (1960), Townes (1969) and stored in 95% ethanol at beeswax; these were buried 2 cm below the soil - 20 °C for further molecular or morphological surface haphazardly under the drip line of a red oak identification. tree as had been done for previous studies (Broadley et al. 2018; Whited 2007). The placement depth was Monthly and annual parasitism rate estimates chosen to mimic natural pupa depths (East 1974; Embree 1965; Holliday 1977). All pupae in each To calculate monthly and annual mortality from deployment were the same age; all had pupated at the parasitism, our estimate of parasitism included pupae end of May from larvae collected mid-May. In total that had developing wasps (larvae and pharate adults) 12,420 winter moth pupae were deployed. and pupae with wasp emergence holes (as shown in To evaluate the effect of native parasitoids on C. Fig. S1). The proportion of pupae parasitized by albicans puparia (winter moth pupae that were para- Pimpla wasps for each deployment was calculated by sitized by C. albicans), in 2014–2016 C. albicans dividing the total number of parasitized pupae by the puparia were also deployed at a subset of sites total number of intact pupae retrieved from the field (Table 1). C. albicans eggs are laid on the surface of (i.e., number of pupae retrieved excluding the number defoliated leaves, and winter moth can become of pupae that were lost due to predation). This method parasitized by C. albicans if they inadvertently of calculating percent parasitism incorporates the fact consume a C. albicans egg while feeding. The C. that parasitism rates can be obscured by predation albicans hatches when inside its host, and when the rates because predation typically occurs on the pupae host winter moth pupates in the soil, the fly larva whether or not they were parasitized. This method develops and forms a puparium inside the host pupa. aims to estimate the true underlying mortality rate of Only one puparium can form inside each host (Wylie each source in the system (Buonaccorsi and Elkinton 1960). The fly puparium is inside the winter moth 1990; Elkinton et al. 1996; Royama 1981; Van cocoon and is readily visible by mid- or late June. For Driesche 1983). To test for any relationship between this study, a total of 3400 C. albicans puparia were the rates of Pimpla wasp parasitism and predation, we deployed. regressed the proportion parasitized by the proportion lost to predation and no trend was detected (Fig. S2); Pupal dissections, incubation, and parasitoid this further suggests that predators do not discriminate collection between parasitized and unparasitized pupae. To calculate the standardized mortality from par-

After each 3- or 6-week pupal deployment, we asitism for each deployment over 31 days (S31), we retrieved the sentinel pupae and characterized their used the following equation: fate as alive (intact) or dead (consumed by predators, 1=n 31 parasitized, or diseased). Without destructive dissec- S31 ¼½ðSpÞ Š tion, it is not possible to determine if the retrieved where Sp is the pupal survivorship from parasitism as a pupae may have a developing endoparasitoid. Thus, proportion and n is the true number of days the pupae the intact pupae were stored in an incubator (Percival were deployed (which ranged from 19 to 45 days with Scientific) until the following spring to allow any 123 naiewne ohi h otesenUie tts2875 States United northeastern the in moth winter Invasive

Table 1 Percent parasitism by Pimpla for each study site and each year of study. Parasitism values are cumulative pupal-stage parasitism rates. The sample size is included in parentheses Site Latitude, longitude Winter moth pupae C. albicans puparia 2014 2015 2016 2017 2014 2015 2016

A. Co-op Extension, Hanson, MA 42.048889, - 70.873806 00.0 (266) 24.2 (102) 11.5 (76) 15.1 (200) – 48.2 (76) 00.0 (97) B. Maquan St., Hanson, MA 42.060694, - 70.844167 20.8 (204) 41.1 (120) 24.6 (159) 31.8 (171) 11.4 (65) 12.6 (93) 00.0 (166) C. Pondview Dr., Falmouth, MA 41.626417, - 70.580417 13.4 (366) – 15.2 (107) 3.1 (167) – – 00.0 (150) D. Centennial Park, Wellesley, MA 42.308444, - 71.266778 33.6 (303) 50.2 (156) 23.2 (141) 19.1 (159) 11.1 (116) 44.9 (123) 00.0 (116) E. Wompatuck SP, Hingham, MA 42.208333, - 70.853056 13.3 (229) 39.4 (113) 23.8 (167) 7.3 (188) – 12.0 (146) 00.0 (183) F. Route 6, Yarmouth, MA 41.686167, - 70.287722 1.3 (343) – 43.3 (35) 00.0 (10) – – 00.0 (47) G. Center St, W. Bridgewater, MA 42.020806, - 70.982306 27.5 (331) –––––– H. Bare Cove Park, Hingham, MA 42.238222, - 70.913389 5.3 (221) –––––– I. Cook Rd, Cumberland, RI 42.012278, - 71.421361 – 35.8 (97) 75.2 (119) –––– J. Parkwood, Drive, Kingston, RI 41.475250, - 71.529444 – 38.2 (85) 29.7 (100) ––47.5 (78) – K. Broad St., Pawcatuck, CT 41.376194, - 71.843528 – 42.6 (88) ––––– L. Garden in the Woods, Framingham, MA 42.340833, - 71.427667 – 47.9 (118) –––6.8 (115) – M. Cadwell Forest, Pelham, MA 42.365500, - 72.437667 ––60.1 (53) –––– N. Mt Toby SF, Sunderland, MA 42.502361, - 72.530111 ––91.8 (15) –––– The first eight sites included here (A–H) are long-term study sites and are used for analyses of parasitism across the years of study and by pupal density, while the sites marked in italics were used for an evaluation of Pimpla parasitism within and outside the main winter moth infestation area 123 2876 H. J. Broadley et al. a mean of 31 days). To calculate the cumulative pupal light. The wasps were sprayed with water twice a day stage parasitism (Pc), we used the following pair of and given honey water. To study development time, equations: the wasps were given access to 100–200 unparasitized ÀÁwinter moth pupae, which were replaced every Sc ¼ Sp1 Â Sp2 Â Sp3 Â Sp4 Â Sp5 5–6 days for a total of 12 rounds between 15 August and and 17 October. Oviposition was monitored during the first hour of exposure to new pupae. When a female Pc ¼ 1 À Sc oviposited into a pupa, we moved the pupa singly to a tube (15 ml Falcon centrifuge tube with a ventilation where Sc is the cumulative pupal stage survivorship hole). The exposed pupae were stored at room from parasitism and Sp1 to Sp5 are the proportion survival from parasitism for successive deployments temperature in ambient light and monitored for for a particular site and year. No value was included subsequent wasp emergence. Of these 120 pupae, 42 for deployments four and five when only three wasps emerged. Wasp development time was calcu- deployments were used to span the pupal life stage. lated. The sex of wasps that emerged from the field- deployed pupae and lab colony was noted and sex- Host selection, development, and sex ratio ratio was evaluated by exposure treatment (field or lab pupae), by date of exposure to parasitism, and by In 2016, we deployed an additional set of 2000 winter season of emergence (fall or spring) using binomial moth pupae to study parasitism depth and host GLM. To assess host choice, a subset of 30 wasps was searching behavior of Pimpla wasps. The pupae were given both winter moth pupae and C. albicans puparia spread one layer thick across the base of wire mesh in a choice test. We ran three trials in September, each cages (mesh size: 6.4 mm sides with a 17 mm mesh with an equal number of winter moth pupae and C. lid) to keep predators out. Two cages each were albicans puparia ranging from 100 to 200 pupae for deployed in two of the study sites, Wellesley and each pupa type. We also compared the monthly Hanson, Massachusetts (Table 1) for three consecu- parasitism rates from pupae and puparia deployed in tive rounds for 35 days each (deployment 1: 7 July–8 the main study plots that received C. albicans puparia August; deployment 2: 8 August–9 September; (Table 1, sites A–F, 2014–2016). We analyzed both deployment 3: 9 September–11 October). Half of the monthly and cumulative pupal stage parasitism by pupae were deployed with a thin layer of leaves over pupal type (winter moth pupae or parasitized by C. top, and the other half had 2 cm of soil then leaves albicans), year, site, and deployment using a GLM over top. After deployment, the retrieved pupae were with a quasibinomial fit. sifted from the soil and stored at room temperature with natural light cycling to allow wasps to complete Parasitism seasonality and year-to-year variation their development and emerge. Every 3 days, the pupae were sprayed with the water-sodium propionate To assess seasonality of parasitism rates, we used a solution, and wasp emergence was recorded. We logistic regression to analyze the monthly rate of tested for an effect of soil and leaf coverage on the parasitism on winter moth pupae weighted by the total emergence counts using a generalized linear model pupae analyzed against the main effect of deployment (GLM) with a negative binomial fit. date and included year and site. We used data from the To study wasp development time, sex ratio, and main study sites (Table 1, sites A–H). Similarly, to host choice, Pimpla wasps that emerged from field- assess seasonality of parasitism, we used a negative deployed pupae in 2016 were reared in the laboratory binomial GLM to evaluate counts of adult wasps that in cages (BugDorm 4F4545 Insect Rearing Cage). emerged from the additional pupae deployed in 2016 Dead wasps were replaced, and newly emerged wasps (‘‘Pimpla host selection, development time, and sex were added, but we always kept 30 wasps per cage ratio’’ section) by date of exposure to parasitism. To with an equal number of males and females. At night, assess pupal parasitism across years, parasitism rates the cages were stored in full dark at 12 °C; during the from the winter moth deployed across all years and day, they were kept at 23 °C and exposed to ambient sites (Table 1, sites A–H) were compared using a logistic model using both our estimation of monthly 123 Invasive winter moth in the northeastern United States 2877 and annual parasitism rates weighted by the total designation of a site as winter moth-infested was pupae analyzed. For our analysis of monthly para- determined from previous studies of winter moth sitism, we included year, site, and deployment as spread (Elkinton et al. 2014, 2015, 2018). predictors. For our analysis of cumulative pupal stage To estimate winter moth pupal density at each long- parasitism, we weighted the logistic regression by the term study plot (Table 1 sites A–H), 16 plastic buckets average number of pupae analyzed across the deploy- (16 cm width 9 28 cm length 9 10 cm height) filled ments and used year and site as predictors. 3 cm deep with sifted peat moss and rainwater drainage holes were placed under each study tree in Parasitism with winter moth spread and density late May before pre-pupal winter moth caterpillars began to spin down from the tree canopies. Each To compare Pimpla wasp parasitism on winter moth bucket was placed at a randomly selected distance pupae between sites infested by winter moth (the first between the tree stem and the edge of the tree canopy eight sites listed in Table 1, to the east of the light along one of eight evenly spaced directions. To test for dotted line in Fig. 1) compared to sites on the edge of density-dependence, we analyzed Pimpla wasp para- the infestation or outside the infestation area (the last sitism by winter moth pupae density using a logistic six sites listed in Table 1, to the west of the light dotted regression of monthly parasitism rates weighted by the line in Fig. 1), we used a logistic regression to analyze total pupae analyzed regressed against the correspond- the monthly rate of parasitism weighted by the total ing density of winter moth pupae (log-transformed) pupae analyzed with infestation status, year, and site with site, year, and deployment as predictors. We also as predictors. We also analyzed the cumulative rate of analyzed the cumulative pupal stage parasitism parasitism weighted by the average number of pupae. against the log-transformed pupal density with year Only years 2015 and 2016 were used for these and site effects. comparisons as they included both heavily infested sites and sites outside the heavily infested area. The

Fig. 1 Average (2014–2017) percent parasitism by Pimpla on sites (A–F) indicated by the gray boxes. The area to the right of winter moth pupae across the pupal deployment sites. The letters the dashed lines approximates the winter moth infestation area for each study site correspond to Table 1 with the six main study for 2007 and 2014 (Elkinton et al. 2014, 2015) 123 2878 H. J. Broadley et al.

Statistical analyses of lepidopterans in multiple families, including geometrids (Yu et al. 2012). We also compared our All analyses were performed in R 3.4.4 (RCoreTeam specimens to the lectotype for Pimpla aequalis 2013) using RStudio, version 1.1.442 (RstudioTeam Provancher from the Universite´ Laval, Quebec City, 2015). For each analysis, the full model was run Quebec, Canada (ULQC). Vouchers for parasitoid initially (including site, year, and deployment effects, species in this research are deposited in the University etc.), the model was evaluated for evidence skew in the of Massachusetts Insect Collection, Amherst, MA residuals or outliers, and any insignificant predictors (UMEC). were dropped sequentially until the best fit model was selected using AIC comparisons. We checked for Molecular comparative analyses overdispersion, and when evidence of overdispersion was noted, we applied a quasibinomial or quasipoisson DNA extraction, amplification, and sequencing distribution. Quasibinomial and quasipoisson analyses do not generate AIC values; thus, to select the best fit A subset of the Pimpla wasps that emerged or were model, we compared the residual deviance of the fit dissected from winter moth or C. albicans pupae were model to that of the null model. A pseudo-R2 was selected for molecular analyses. When possible, three calculated by comparing the residual deviance of the adult samples and one larval sample were selected for fit model against the null model (deviance null each location and study year; otherwise, up to four model - deviance fit model/deviance null model). wasps of any life stage were selected for a total of 77 All graphical data were displayed using ggplot2 field-collected individuals and 20 laboratory-reared (Wickham 2009). wasps (Table S1). DNA was extracted using the QIAGEN DNeasy Blood and Tissue Kits following Morphological comparative analyses the company protocol with the following modifica- tions: for larvae, individuals were destructively sam- Following initial identification of our specimens as pled by grinding with a mortar and pestle; for adults, belonging to the genus Pimpla, further morphological DNA was extracted from a single leg removed from and molecular identification was conducted using a the specimen; for both life stages, DNA was eluted subset of 302 samples (289 from winter moth and 13 twice in 100 ll Buffer AE instead of once with 200 ll. from C. albicans puparia). This subset included wasps All DNA extractions were stored at - 20 °C for reared from all collection sites, seasons, and years, and subsequent analysis. included males and females. The specimens were A portion of the mitochondrial locus cytochrome c initially sorted into putative species based on mor- oxidase subunit I (COI) was amplified using standard phology; species identification was attempted using PCR techniques. For a subset of individuals collected keys and diagnostic information in Townes (1940) and across years, sites, life stages, and life history Townes et al. (1960). Specimens were also compared (Table S1) fragments from three additional nuclear with authoritatively determined specimens of Pimpla gene regions were amplified: the carbomoylphosphate turionellae L., Pimpla contemplator Mu¨ller, Pimpla synthase domain (Cadherin, rudimentary, CAD), hesperus Townes, Pimpla aquilonia Cresson, and elongation factor 1-a (EF1-a), and the D2 and D3 Pimpla disparis Viereck in the Smithsonian Institution expansion segments of the large subunit ribosomal National Museum of Natural History, Washington, RNA gene (28S). The primers and temperature DC (USNM). All but the last two species have been profiles used are outlined in Table S2. For each locus, recorded as attacking winter moth (Silvestri 1941; a master mix was prepared using the following ratios Wylie 1960; Humble 1985). Pimpla flavicoxis has of reagents per sample: 17.3 ll nuclease free water, been reported from winter moth in the British Isles and 0.5 ll dNTPs, 5 ll59 GoTaq Buffer (Promega), was previously treated as a junior synonym of 0.2 ll GoTaq (Promega), 0.5 ll of both the forward P. aquilonia. However, there are no specimens at the and reverse primer (10 lM each), and 1 ll of eluted USNM identified as P. flavicoxis. P. disparis was DNA. PCR reactions were run on a BioRad T100 introduced into Canada and the U.S. to control other thermocycler, and the resulting PCR products were lepidopteran pests. It has been reported as a parasitoid visualized on a 1.5% agarose gel stained with 123 Invasive winter moth in the northeastern United States 2879

SYBERsafe (Invitrogen, Carlsbad, CA) to verify sequences included 57 Pimpla specimens representing amplification. Samples that produced bands of the 13 of the 19 extant described species in the Nearctic expected fragment size for each locus were then Region (Yu et al. 2012). When no representative cleaned prior to sequencing using Exonuclease 1 sequence was available for a particular Pimpla species (ThermoScientific) and Thermolabile Recombinant by the Canadian National Collection, we searched Shrimp Alkaline Phosphatase (New England Bio- GenBank for a representative sample, followed by any Labs). The resulting products were submitted to The other sequences available in BOLD. Sequences iden- Yale University DNA Analysis Facility on Science tified as P. aequalis and as Pimpla sp. that were the Hill for Sanger sequencing in both sense and anti- closest matches in GenBank to sequences of our each sense orientations. of our two Pimpla clades were included; these were The resulting sequences were visualized, and the associated with publications (Carpenter and Wheeler forward and reverse sequences aligned and edited 1999; Hebert et al. 2016) and from the International using Geneious R8.1.8 and R9 (Biomatters Ltd.). The Barcode of Life and NCBI GenBank. ends of the aligned sequences were trimmed by hand JModelTest was used to select the best substitution to remove primer sequences and so that all sequences model for nucleotide evolution, as implemented in the had a high-quality score ([ 90% HQ nucleotide CIPRES Science Gateway (Miller et al. 2010). We reads). The presence of heterozygous sites was performed neighbor-joining, maximum likelihood, determined by Geneious and encoded using IUPAC- and Bayesian reconstructions using the GTR substi- IUB ambiguity codes. All ambiguous regions were tution model. Neighbor-joining analyses were run in subsequently inspected by eye. For our COI fragment Geneious using 1000 bootstrap replications and a sequences, we looked for evidence of nuclear mito- majority rule (50%) consensus threshold. Maximum chondrial DNAs (NUMTs) or pseudogenes by exam- likelihood analyses were run using PhyML (Guindon ining for the presence of stop codons based on et al. 2010) with 100 bootstrap replications. Bayesian translation with Invertebrate Mitochondrial DNA analyses were run using MrBayes 3.2.6 (Huelsenbeck genetic code. and Ronquist 2001) with a MCMC chain length of 1,000,000 and a burn in length of 10%. The resulting Phylogenetic analysis gene trees were then visualized using FigTree Version 1.4.2 (Rambaut 2014). If any of our sequences were identical, we only To determine whether specimens identified as P. included a single representative haplotype. The num- aequalis might be members of a cryptic species ber of samples and sample identification for each complex, we used a multilocus genealogical concor- haplotype is outlined in Table S3. To compare our dance approach (Andersen et al. 2010; Dettman et al. Pimpla COI sequences to previously published Pimpla 2006; Groeneveld et al. 2009; Starrett and Hedin 2007) COI sequences, we searched the National Center for to estimate the number of species present in our Biotechnology Information (NCBI) GenBank data- dataset. This method considers lineage sorting in base and the University of Guelph Centre for Biodi- multiple, independent loci and has become a common versity Genomics’s Barcode of Life Data Systems approach for species delineation. For these analyses, (BOLD). We initially downloaded triplicate we created separate alignments for each gene fragment sequences from each Pimpla species available across including each specimen from which all target loci the two repositories, but when the triplicates were were successfully amplified. In addition, we included identical to each other or nearly identical ([ 99% publicly available sequences from Labena grallator identical), we then retained one representative (Say) (Hymenoptera: Ichneumonidae) as the outgroup sequence for each Pimpla species available. From for each alignment. Individual gene trees were the BOLD sequences, we prioritized sequences estimated for each locus, and the congruence of the acquired from Pimpla samples in the hymenopteran topologies of the reconstructed gene trees were then collection of the Canadian Natural Collection of visualized by inferring a majority-rule consensus tree Insects, Arachnids and Nematodes (Agriculture and using PAUP (Swofford 2003). Agri-Food Canada) accessed by A. Bennett (accession numbers start with ‘BOLD HYCNG’). These 123 2880 H. J. Broadley et al.

Results Parasitism with winter moth spread and density

Parasitoid collection Pupal density of the study sites had a significant effect on the monthly (df = 62, pseudoR2 = 0.25, p = 0.039) Of the 6580 retrieved pupae and puparia that escaped and cumulative pupal stage parasitism (df = 19, predation (5009 winter moth pupae and 1571 C. pseudoR2 = 0.39, p = 0.036), with a significant effect albicans puparia, Table 1) over the study period of year in both models (Fig. 2). No significant (2014–2017), 342 were parasitized by Pimpla wasps difference was found in percent Pimpla wasp para- (305 winter moth pupae and 37 C. albicans puparia). sitism on winter moth pupae that were deployed in Besides the Pimpla wasps, we recovered only two sites on the edge of the current winter moth infestation other species: two winter moth pupae (Wellesley, MA area compared to the heavy infestation area (Fig. 1). and Pawckatuck, CT; 24 June–5 August 2015) were However, there was a trend toward higher parasitism parasitized by a species of , and four C. rates at the edge of the winter moth infestation albicans puparia (Kingston, RI; 5 August–18 Septem- (Fig. S5). Pupae deployed in sites at the edge of the ber 2015) were parasitized by a brood of diapriid winter moth infestation had a mean cumulative pupal wasps. Of the Pimpla wasps recovered from field- stage parasitism rate of 0.53 ± 0.08 (mean ± SE), deployed sentinel pupae and puparia, 46% were adults while the cumulative parasitism was 0.32 ± 0.03 in and the rest larvae. the infested sites.

Parasitism seasonality, overwintering, and year-to- Wasp development time and sex ratio year variation It took 21.2 ± 0.6 SE days from the date of parasitism Monthly parasitism on winter moth pupae varied from to adult wasp emergence (n = 39) for pupae para- 0 to 52% (Fig. S3). Deployment was marginally sitized in the laboratory. No wasp emergence was significant with parasitism rates from early August to noted after 9 October until spring. Thus, if Pimpla mid-September slighter higher than those of either late wasps parasitize between 1 June and 1 October June to early August or mid-September to late October (122 days) and if wasps take 21 days from oviposition (df = 100, pseudoR2 = 0.34, p = 0.06). Pupae to emergence, then we can expect up to 5 generations exposed to Pimpla wasps after the first week of per season and a final 6th overwintering generation. August had wasps that did not emerge until the This estimate is from individuals held at a constant following spring (the overwintering generation), and 23 °C temperature; however, in the field average most of the overwintering wasps (91%) were from temperatures are slightly cooler (* 19 °C, Table S4), winter moth pupae that were exposed to wasps after so development may be slower under field conditions. the first week of September. Cumulative pupal stage Across studies (field deployed pupae and laboratory parasitism (one minus the product of the survivorship rearing), there were 305 Pimpla females and 238 of each pupal deployment for the duration of the pupal Pimpla males, which suggests a 1:1 female to male life stage, see ‘‘Monthly and annual parasitism rate ratio. estimates’’ section) ranged from 0 to 92% (Table 1). Pimpla wasps were recovered from all years and sites, Pimpla host selection though some sites had a year without recoveries. There was a significant effect of year but not site or From the three deployments of pupae placed in deployment date when analyzing monthly parasitism Wellesley, MA and Hanson, MA in 2016, there was rates (df = 57, pseudoR2 = 0.40, p = 0.034), and site a significant effect of soil treatment (df = 8, pseu- was significant when analyzing the cumulative pupal doR2 = 0.57, p = 0.0043) and deployment stage parasitism rates (Fig. S4; df = 13, pseudoR2- (p = 0.0018). More wasps emerged from the pupae = 0.72, p = 0.009). Parasitism rates were highest in that were covered only by a layer of leaves than from 2015. those buried under soil and leaves. In the host choice study, we did not observe any wasps attempting to oviposit in the C. albicans puparia, and no wasps 123 Invasive winter moth in the northeastern United States 2881

Fig. 2 Logistic relationship between monthly (left) and cumu- point indicates each site for each year, the solid lines show the fit lative pupal stage (right) Pimpla parasitism on pupae by winter model, and the dashed lines show confidence intervals moth pupal density across sites for the six main study sites. Each emerged from these trials. However, from the field Pimpla sp. 2 are presented in Fig. 5a–f, respectively studies, we found that C. albicans puparia can be (note: we did not rear females of Pimpla sp. 2). While parasitized by Pimpla spp. but at a significantly lower there appear to be subtle morphological differences rate than winter moth; this was true for the model that between Pimpla sp. 1 and Pimpla sp. 2, particularly included monthly parasitism rates (df = 83, pseudoR2 shape of the mesosoma, whether they are reliable for = 0.39, p = 0.00012) and the model with cumulative differentiating these species is equivocal because we pupal stage parasitism rates (df = 22, pseudoR2 have only one adult male of Pimpla sp. 2. All other = 0.64, p = 0.0018). Mean cumulative parasitism by specimens of Pimpla sp. 2 were larvae from para- Pimpla on C. albicans puparia was 0.15 as compared sitized hosts. to 0.27 for winter moth pupae. We acquired high quality COI sequences for 74 individuals representing all sites and years (50 adults Molecular and morphological comparative and 24 larvae; Table S1). Based on reconstruction of analyses the phylogeny using the COI gene fragment, our samples separated into two distinct clades that exhib- Based on morphological features, a subset of Pimpla ited 9.7–10.1% sequence divergence (Fig. 3). All wasps were identified by Dr. David Wahl (Utah State nucleotide differences between the two clades repre- University) as P. aequalis. Subsequently, additional sented third-codon substitutions and thus likely rep- samples were identified by the second author (RRK) as resent genetic differences accumulated due to genetic P. aequalis using the key presented in Townes (1940) drift and not selection. Both Pimpla clades included and Townes et al. (1960) and comparison with the wasps acting as primary parasitoids and hyperpara- lectotype for P. aequalis (a female, Fig. S7). The sitoids (49 primary and 26 hyperparasitoids; identification was confirmed by another ichneumonid Table S1). Additionally, we obtained high quality systematist (B. Santos, Smithsonian Institution sequences for fragments of CAD, EF1-a, and 28S NMNH) through examination of a subset of the from 26 Pimpla wasps (19 Pimpla sp. 1 individuals identified specimens. However, results from analysis and 7 Pimpla sp. 2 individuals as from the COI of molecular data (see below) suggest that the Pimpla analyses). For CAD, sequences from the two Pimpla wasps in this study actually comprise two species clades were 1.5–3.4% different from each other, with hereafter referred to as Pimpla sp. 1 and Pimpla sp. 2. six base-pair differences fixed between the two Pimpla Images of a female of Pimpla sp. 1 and a male of clades (Fig. 4, Table S7). Similarly, for EF1-a the two

123 2882 H. J. Broadley et al.

Fig. 3 Phylogentic inference of our Pimpla samples (bolded) were collapsed as outlined in Table S2. Pimpla aequalis sp. 1 are and representative sequences. Representative sequences were noted with red and sp. 2 with blue. Branch lengths are drawn downloaded from NCBI GenBank and BOLD. The tree was proportional to the rate of change observed. The number to the constructed using a Bayesian analysis with a 604 bp region of left each node represents the bootstrap support value for the the COI locus. Where our sequences were identical, the groups branch (Bayesian over the Maximum Likelihood)

123 Invasive winter moth in the northeastern United States 2883

Fig. 4 Phylogentic inferences using a COI, b CAD, and c EF1- rate of change observed. The number to the left each node a gene regions for only our P. aequalis sp. 1 (red) and sp. 2 represents the bootstrap support value for the branch (Bayesian (blue) samples. Branch lengths are drawn proportional to the over the Maximum Likelihood) clades were between 0.7 and 1.6% different with 3 the P. aequalis lectotype is a female. Determining the base-pairs consistently different between clades. See identities of these clades, and discerning the identities Tables S6–S8 for distance matrices for each align- of Holarctic species of Pimpla in general, would ment. For 28S, there were no differences between require extensive analysis of morphological and individuals from the two COI clades, with all but two molecular data for Pimpla species in the Nearctic individuals having identical sequences; the two indi- and Palearctic regions, including primary type spec- viduals differed from the other by a single base-pair imens. The sequences generated from our specimens substitution. Because 28S was invariant, it was left out did not match any published sequences available for of the multilocus analyses. The trees constructed from other Pimpla species, which includes 13 of the 19 each of these three loci (Fig. 4) and the majority rule extant described species in the Nearctic Region (Yu consensus tree (Fig. S6) all supported the presence in et al. 2012) and an additional six species of Pimpla our samples of two cryptic Pimpla species within what from outside the Nearctic Region. The only molecular we considered P. aequalis based on morphology matches were for other unknown Pimpla species (Fig. 5). collected from the northeastern United States and Based on these comparisons, we appear to have two southeastern Canada. Thus, Pimpla sp. 1 and Pimpla distinct species of Pimpla that fit P. aequalis sensu sp. 2 are presumably native to North America. Townes (1940) and Townes et al. (1960), and these species are not any Palearctic species available to us known to attack winter moth in Europe. It is likely that Discussion either Pimpla sp. 1 or Pimpla sp. 2 is P. aequalis. The lectotype of P. aequalis is more similar morpholog- Biotic resistance is the process by which native natural ically to specimens of Pimpla sp. 1 than Pimpla sp. 2; enemies spill over from native species to attack an however, Pimpla sp. 2 is known from a male only, and invasive species and reduce the success of that invader

123 2884 H. J. Broadley et al.

Fig. 5 The two Pimpla species reared from winter moth pupae f Pimpla sp. 2, male, d Lateral habitus, e lateral of head and in this study. a–c Pimpla sp. 1, female a Lateral habitus; b lateral mesosoma, f Dorsal of head and mesosoma. Scale of head and mesosoma, c Dorsal of head and mesosoma, d– bars = 1.00 mm

(Diez et al. 2008; Elton 1958; Levine et al. 2004; Sakai related unknown species. While the association of et al. 2001; Shea and Chesson 2002). We found these Pimpla wasps with winter moth is recent, their evidence of biotic resistance to invasive winter moth impact is notable; estimates of parasitism across our populations in the northeastern United States; in this study sites and years were on average 6% and were most recent invasion, winter moth populations are found to be as high as 52% on winter moth pupae in sustaining heavy, density dependent parasitism by two infested areas. Pimpla wasps were found across all of cryptic species of Pimpla. One of the Pimpla wasps is our study plots. Where winter moth has invaded, likely P. aequalis, while the other appears to be a Pimpla wasp parasitism responded to winter moth

123 Invasive winter moth in the northeastern United States 2885 pupal density in a positive density dependent manner the area. We found that Pimpla wasps exhibited and thus has the potential to be regulatory. particularly high parasitism 95 km beyond the nearest The ability of a native predator or parasitoid to high density winter moth infestation (i.e., Framing- respond functionally or numerically to a primary host ham, MA) and 22 km from the nearest capture of population, while also using an alternative host when winter moths in pheromone traps (i.e., Orange, MA). the primary host is at low densities, may result in The finding that Pimpla wasps were present across the particularly effective suppression of populations of study area, including areas beyond the range that non-native invader by a native species (Nechols et al. winter moth has spread, supports the hypothesis that 1992; Shea and Chesson 2002). In this way, the native the Pimpla species are native. Pimpla wasp population natural enemy has the ability to build up using the densities respond to, but do not depend on, the novel host but can also use other host species when this presence of winter moth. In fact, we found particular new host is less abundant. As a result, the natural high parasitism rates at the leading edge of the winter enemy is maintained in the community and can moth infestation. This is likely because in these aggregate or respond numerically when the invasive interior sites Pimpla wasps, as native generalist alien population outbreaks or spreads. This pattern of parasitoids, are maintained at high densities by a host use is common for a number of generalist robust Lepidoptera community. In this way, Pimpla predators, parasitoids, and pathogens (deRivera et al. wasps may provide a biotic resistance barrier to the 2005; Hassell and Rogers 1972; Holling 1973; Mur- spread of winter moth, which was found by Elkinton doch 1969; Oaten and Murdoch 1975; Schenk and et al. (2014) to be slowing. Pimpla wasps may help Bacher 2002; Strauss et al. 2012). Parasitism by control winter moth spread, but this warrants further Pimpla wasps in this study exhibited these character- study. istics. While Pimpla wasps may build up using winter If Pimpla wasps act as hyperparasitoids (a para- moth as a host species and respond to high densities sitoid of a parasitoid) of any native or introduced with higher rates of parasitism, it also appears to be parasitoids (biological control agents), then they have able to use other host species when winter moth the potential to reduce population control of the populations are not at high densities. invasive species by inflicting more mortality on the We detected a positive density dependent relation- biological control agent than on the invasive species ship of parasitism to winter moth density, suggesting itself. However, if a parasitoid inflicts more mortality that Pimpla wasps may have a regulatory effect on the on the invasive species than on any introduced winter moth population densities. Further, since the biological control agent, then it aids in controlling Pimpla wasps are multivoltine, the wasp population the pest population (Brodeur 2000; Nechols et al. may be able to build up in outbreak populations in a 1992; Sullivan 1987). While we found that Pimpla numerical response, which could control an outbreak- wasps can hyperparasitize, as is typical of many ing population (Holling 1973). However, we did not pimpline wasps (Bennett 2008; Fitton et al. 1988) and detect increased parasitism rates over either season or was found previously (Broadley et al. 2018), they years and were not able to test for a numerical response appear to do so only facultatively. From our laboratory to densities. In the absence of a documented numerical host range study, when given a choice of winter moth response, it may be that the density dependent pupae or winter moth pupae parasitized by C. response we found may arise from a Type III albicans, the wasps only parasitized winter moth functional response driven by Pimpla wasp host pupae. From our field study, both species of Pimpla switching behavior (Holling 1959; Murdoch 1969). wasps can act as hyperparasitoids, but they do so at While pimplines typically have a preferred host rates significantly lower than their rate of primary species (niche specialization), they are facultative parasitism. Hyperparasitism on C. albicans at these generalists and can use a broad host range spanning sites was primarily caused by wasps from the genus multiple lepidopteran families (Bennett 2008; Fitton Phygadeuon rather than Pimpla (Broadley et al. 2018). et al. 1988; Krombein et al. 1979). Parasitism of winter Together, this further demonstrates that Pimpla wasps moth by Pimpla wasps is likely the result of natural contribute to the population control of winter moth. enemy spillover from the native congener Bruce We detected two species of Pimpla parasitizing spanworm or other related lepidopteran species in winter moth pupae; one is likely P. aequalis, the other 123 2886 H. J. Broadley et al. is unknown and is possibly an undescribed species. other morphologically similar congeners, were found For our estimates of percent parasitism, we did not there (Morin 2015); however, the study did not include distinguish between the two species of Pimpla, but sentinel winter moth and thus showed co-occurrence from the molecular work, one clade is better repre- but not parasitism. sented than the other; the majority (88%) of the We were surprised that Pimpla sp. 1 detected in this randomized samples we tested belonged to Pimpla sp. study seemed to be the only major species that 1. Both species were found across study site, season, parasitized winter moth pupae. The lack of additional and year, and both species acted as both primary and parasitoids and slow recruitment of Pimpla wasps may hyperparasitoids. While both species were initially help explain why winter moth has been such a high considered P. aequalis based on morphology, we were density pest in its introduced region. We looked for not able to acquire DNA from the lectotype to discern parasitism by P. contemplator and P. turionellae, if one of our two species is P. aequalis. Pimpla which are known parasitoids of winter moth in Europe aequalis is known only from North America, and we (Wylie 1960) and were introduced to southeastern consider both Pimpla species we recovered from Canada in an attempt to control winter moth, but there winter moth as native to North America. However, is no evidence that they established (Carlson 1979; some species of Pimpla are known to have a Holarctic Graham 1958). Further, P. disparis, a parasitoid distribution (Yu et al. 2012). The overall geographic introduced in this region to control gypsy moth, was distribution of the Pimpla species reported here cannot also a potential candidate since P. disparis has a broad be determined until their identities have been dis- host range, attacking lepidopterans of at least 14 cerned; this would require more extensive taxonomic families (Schaefer et al. 1989). However, these Pimpla research on Pimpla species in the Nearctic and species were not detected. Besides the two species of Palearctic regions, such as a revision that includes Pimpla reported here, we only had a few cases of both morphological and molecular data. parasitism by diapriid and Cratichneumon wasps. Winter moth has been extensively studied in its Cratichneumon culex (Muell.) has been recorded as an native range in Europe (e.g. Klemola et al. 2008; important parasitoid of winter moth pupae in Europe Myers and Cory 2013; Tenow et al. 2013; Varley et al. (East 1974; Hassell 1969; Varley et al. 1973; Wylie 1973; Vindstad et al. 2011, 2013), as well as in the 1960), and an undescribed Cratichneumon species was prior accidental introductions to North America (e.g. reared from winter moth in British Columbia (Humble Embree 1965; Roland 1990; Roland and Embree 1985). As far as we know, C. culex has not been 1995); however, this is the first report of Pimpla introduced to North America (Embree 1966; Graham species as important parasitoids of winter moth. To our 1958). knowledge none of the studies of winter moth pupal mortality in Nova Scotia recorded parasitism by Pimpla wasps, although native parasitoids were noted Conclusions (Embree 1965; Graham 1958; Macphee et al. 1988; Pearsall and Walde 1994; Roland 1990). In British Our findings suggest an important role of Pimpla Columbia, Coccygomimus (= Pimpla) hesperus was wasps in the population dynamics of winter moth, an recorded from Operophtera spp. pupae (Humble invasive forest pest. Overall, this study shows that 1985) but was not recorded in later studies (Roland biotic resistance from a native parasitoid is affecting 1990; Roland and Embree 1995). This suggests that the dynamics of the winter moth invasion. We urge the wasps were accidently overlooked in the prior future research on parasitism by native species in the studies, that Pimpla wasps now show more host study of biological invasions, as knowledge of biotic switching to winter moth than when winter moth was resistance from native natural enemies on invasive first introduced to North America in the 1930s, or that populations is essential to our overall understanding of the region of this study has more abundant Pimpla invasions. We also encourage insect biological control wasp populations with a wider host range than was practitioners to consider not only the effect of an found in Nova Scotia or British Columbia. Surveys introduced biocontrol agent but also the effect of were conducted in winter moth infested sites in coastal native parasitoids and the potential interactions Maine, and species of Pimpla, likely P. aequalis and 123 Invasive winter moth in the northeastern United States 2887 between introduced biocontrol agents and native Buonaccorsi JP, Elkinton JS (1990) Estimation of contempo- natural enemies. raneous mortality factors. Res Popul Ecol 32:151–171 Bu¨rgi LP, Roltsch WJ, Mills NJ (2015) Allee effects and pop- ulation regulation: a test for biotic resistance against an Acknowledgements The authors thank R. Crandall, N. invasive leafroller by resident parasitoids. Popul Ecol Ayres, A. Roehrig, T. Dowling, N. Milano, Q. Dupupet, R. 57(1):215–225. https://doi.org/10.1007/s10144-014-0451- Hennessy, C. Camp, E. Mooshian, K. Donahue, E. Lee, J. Cox, 4 E. Kelly, and G. Greenberg, D. Swanson, D. Adams, R. Callaway RM, Montesinos D, Williams K, Maron JL (2013) Casagrande, E. Amore, and T. Peckham for their assistance with Native congeners provide biotic resistance to invasive this research. We are grateful to B. Santos (USNM) and A. Potentilla through soil biota. Ecology 94:1223–1229 Bennett (CNC) for advice on ichneumonid identification and Carlson RW (1979) Family Ichneumonidae. In: Krombein KV, . We thank T. Litwak (Systematic Entomology Hurd PD, Smith DR, Burks BD (eds) Catalog of Hyme- Laboratory, USDA-ARS) for imaging the Pimpla specimens. noptera in America north of Mexico. Smithsonian Institu- We thank S. Klopfstein and J. Mottern for their suggestions on tion Press, Washington, pp 315–741 primers and DNA amplification. Lastly, we thank Drs. Van Carpenter JM, Wheeler WC (1999) Towards simultaneous Driesche, Adler, Normark, and Burand for comments on earlier analysis of morphological and molecular data in Hyme- versions of the manuscript. This material is based on work noptera. Zool Scr 28:251–260 supported by Cooperative Agreements from the USDA APHIS Carrillo-Gavilan A, Moreira X, Zas R, Vila M, Sampedro L [12 13 14-8225-0464-CA] and from the USDA Forest Service (2012) Early resistance of alien and native pines against [13-CA-1140004-236-CA] as well as a Summer Research two native generalist insect herbivores: no support for the Scholarship from the University of Massachusetts Amherst natural enemy hypothesis. Funct Ecol 26:283–293 Natural History Collections and an Irwin Martin Award from the Dearborn KW, Heard SB, Sweeney J, Pureswaran DS (2016) Graduate Program in Organismic and Evolutionary Biology, Displacement of cinnamopterum (Coleoptera: University of Massachusetts Amherst. Mention of trade names Cerambycidae) by its Invasive Congener Tetropium fus- or commercial products in this publication is solely for the cum. Environ Entomol 45:848–854 purpose of providing specific information and does not imply deRivera CE, Ruiz GM, Hines AH, Jivoff P (2005) Biotic recommendation or endorsement by the USDA. USDA is an resistance to invasion: native predator limits abundance equal opportunity provider and employer. and distribution of an introduced crab. Ecology 86:3364–3376 Compliance with ethical standards Dettman JR, Jacobson DJ, Taylor JW (2006) Multilocus sequence data reveal extensive phylogenetic species Conflict of interest The authors declare that they have no diversity within the Neurospora discreta complex. conflict of interest. Mycologia 98:436–446 Diez JM, Sullivan JJ, Hulme PE, Edwards G, Duncan RP (2008) Darwin’s naturalization conundrum: dissecting taxonomic patterns of species invasions. Ecol Lett 11:674–681 Donahue KL, Broadley HJ, Elkinton JS, Burand JP, Huang WF, References Andersen JC (2018) Using the SSU, ITS, and ribosomal DNA operon arrangement to characterize two micro- Andersen JC, Gruwell ME, Morse GE, Normark BB (2010) sporidia infecting Bruce spanworm, Operophtera bruceata Cryptic diversity in the Aspidiotus nerii complex in Aus- (Lepidoptera: Geometridae). J Eukaryot Microbiol. https:// tralia. Ann Entomol Soc Am 103:844–854 doi.org/10.1111/jeu.12685 Bennett AMR (2008) Review and identification keys to the Duan JJ, Taylor PB, Fuester RW, Kula RR, Marsh PM (2013) ichneumonid parasitoids (Hymenoptera: Ichneumonidae) Hymenopteran parasitoids attacking the invasive emerald of Nearctic Choristoneura species (Lepidoptera: Tortrici- ash boror (Coleoptera: Buprestidae) in western and central dae). Can Entomol 140:1–47 Pennsylvania. Fla Entomol 96:166–172 Broadley HJ, Boucher M, Burand JP, Elkinton JS (2017) The Duan JJ, Abell KJ, Bauer LS, Gould J, Van Driesche R (2014) phylogenetic relationship and cross-infection of nucle- Natural enemies implicated in the regulation of an invasive opolyhedroviruses between the invasive winter moth pest: a life table analysis of the population dynamics of the (Operophtera brumata) and its native congener, Bruce emerald ash borer. Agric For Entomol 16:406–416 spanworm (O. bruceata). J Invertebr Pathol 143:61–68 East R (1974) Predation on soil-dwelling stages of winter moth Broadley HJ, Kelly EA, Elkinton JS, Kula RR, Boettner GH at Wytham Woods Berkshire. J Anim Ecol 43:611–626 (2018) Identification and impact of hyperparasitoids and Elkinton JS, Healy WM, Buonaccorsi JP, Boettner GH, Hazzard predators affecting Cyzenis albicans (Tachinidae), a AM, Smith HR, Liebhold AM (1996) Interactions among recently introduced biological control agent of winter moth gypsy moths, white-footed mice, and acorns. Ecology (Operophtera brumata L.) in the northeastern U.S.A. Biol 77:2332–2342 Control 121:99–108 Elkinton JS, Boettner GH, Sermac M, Gwiazdowski R, Hunkins Brodeur J (2000) Host specificity and trophic relationships of RR, Callahan J, Scheufele SB, Donahue CP, Porter AH, hyperparasitoids. In: Hochberg ME, Ives AR (eds) Para- Khrimian A, Whited BM, Campbell NK (2010) Survey for sitoid population ecology. Princeton University Press, winter moth (Lepidoptera: Geometridae) in northeastern Princeton North America with pheromone-baited traps and 123 2888 H. J. Broadley et al.

hybridization with the native Bruce spanworm (Lepi- Gwiazdowski RA, Elkinton JS, Dewaard JR, Sremac M (2013) doptera: Geometridae). Ann Entomol Soc Am Phylogeographic Diversity of the Winter Moths Oper- 103:135–145 ophtera brumata and O. bruceata (Lepidoptera: Elkinton JS, Liebhold A, Boettner GH, Sremac M (2014) Geometridae) in Europe and North America. Ann Entomol Invasion spread of Operophtera brumata in northeastern Soc Am 106:143–151 United States and hybridization with O. bruceata. Biol Hassell MP (1969) A study of mortality factors acting upon Invasions 16:2263–2272 Cyzenis albicans (Fall), a tachinid parasite of winter moth Elkinton JS, Boettner GH, Liebhold A, Gwiazdowski R (2015) (Operophtera brumata (L)). J Anim Ecol 38:329–339 Biology, spread, and biological control of winter moth in Hassell MP (2000) Host-parasitoid population dynamics. the eastern United States. In: Team TFHTE (ed). USDA J Anim Ecol 69:543–566 Forest Service Publication, Morgantown Hassell MP, Rogers DJ (1972) Insect parasite responses in the Elkinton JS, Boettner GH, Broadley HJ, Reardon R, Weeks Jr development of population models. J Anim Ecol RD (2018) Biological control of the winter moth in the 41:661–676 northeastern North America. USDA Forest Service Forest Havill NP, Elkinton J, Andersen JC, Hagen SB, Broadley HJ, Health Assessment and Applied Science Team 2018-03:0- Boettner GJ, Caccone A (2017) Asymmetric hybridization 9 between non-native winter moth, Operophtera brumata Elton CS (1958) The ecology of invasions by and plants. (Lepidoptera: Geometridae), and native Bruce spanworm, Methuen, London Operophtera bruceata, in the Northeastern United States, Embree DG (1965) The population dynamics of the winter moth assessed with novel microsatellites and SNPs. Bull Ento- in Nova Scotia, 1954–1962. Mem Entomol Soc Can mol Res 107:241–250 97:5–57 Hebert PDN, Ratnasingham S, Zakharov EV, Telfer AC, Lev- Embree DG (1966) The role of introduced parasites in the esque-Beaudin V, Milton MA, Pedersen S, Jannetta P, control of the winter moth in Nova Scotia. Can Entomol deWaard JR (2016) Counting species with DNA 98:1159–1168 barcodes: Canadian insects. Philos Trans R Soc B Faillace CA, Lorusso NS, Duffy S (2017) Overlooking the 371:20150333 smallest matter: viruses impact biological invasions. Ecol Heimpel G, Mills N (2017) Biological control: ecology and Lett 20:524–538 applications. Cambridge University Press, Cambridge. Fitton MG, Shaw MR, Gauld ID (1988) Pimpline ichneumon- https://doi.org/10.1017/9781139029117 flies: Hymenoptera, Ichneumonidae (Pimplinae). Royal Hobbs RJ, Higgs E, Harris JA (2009) Novel ecosystems: Entomological Society of London, London implications for conservation and restoration. Trends Ecol Garnas JR, Auger-Rozenberg MA, Roques A, Bertelsmeier C, Evol 24:599–605 Wingfield MJ, Saccaggi DL, Roy HE, Slippers B (2016) Holliday NJ (1977) Population ecology of winter moth (Oper- Complex patterns of global spread in invasive insects: eco- ophtera brumata) on apple in relation to larval dispersal evolutionary and management consequences. Biol Inva- and time of bud burst. J Appl Ecol 14:803–813 sions 18:935–952 Holling CS (1959) The components of predation as revealed by Gauld ID (1991) The Ichneumonidae of Costa Rica, 1. Intro- a study of small mammal predation of the European pine duction, keys to subfamilies, and keys to the species of the sawfly. Can Entomol 91:293–320 lower pimpliform subfamilies Rhyssinae, Poemeniinae, Holling CS (1973) Resilience and stability of ecological sys- Acaenitinae and Cylloceriinae. Mem Am Entomol Inst tems. Annu Rev Ecol Syst 4:1–23 47:1–589 Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian Goulet H, Huber JT (1993) Hymenoptera of the world: an inference of phylogeny. Bioinformatics 17:754–755 identification guide to families. Centre for Land and Bio- Humble LM (1985) Final-instar larvae of native pupal parasites logical Resources Research, Ottawa and hyperparasites of Operophtera spp. (Lepidoptera, Grabenweger G, Kehrli P, Zweimuller I, Augustin S, Avtzis N, Geometridae) on southern Vancouver Island. Can Entomol Bacher S, Freise J, Girardoz S, Guichard S, Heitland W, 117:525–534 Lethmayer C, Stolz M, Tomov R, Volter L, Kenis M (2010) Jeschke J, Go´mez Aparicio L, Haider S, Heger T, Lortie C, Temporal and spatial variations in the parasitoid complex Pysˇek P, Strayer D (2012) Support for major hypotheses in of the horse chestnut leafminer during its invasion of invasion biology is uneven and declining. NeoBiota 14:1 Europe. Biol Invasions 12:2797–2813 Keane RM, Crawley MJ (2002) Exotic plant invasions and the Graham AR (1958) Recoveries of introduced species of para- enemy release hypothesis. Trends Ecol Evol 17:164–170 sites of the winter moth, Operophtera brumata (L.) Kenis M, Roy HE, Zindel R, Majerus MEN (2008) Current and (Lepidoptera: Geometridae), in Nova Scotia. Can Entomol potential management strategies against Harmonia axyr- 90:595–596 idis. Biocontrol 53:235–252 Groeneveld LF, Weisrock DW, Rasoloarison RM, Yoder AD, Klemola T, Andersson T, Ruohomaki K (2008) Fecundity of the Kappeler PM (2009) Species delimitation in lemurs: mul- autumnal moth depends on pooled geometrid abundance tiple genetic loci reveal low levels of species diversity in without a time lag: implications for cyclic population the genus Cheirogaleus. BMC Evol Biol 9:30 dynamics. J Anim Ecol 77:597–604 Guindon S, Dufayard JF, Lefort V, Anisimova M, Hordijk W, Krombein KV, Hurd PD, Smith DR, Burks BD (1979) Sym- Gascuel O (2010) New algorithms and methods to estimate phyta and (Parasitica). Catalog of Hymenoptera maximum-likelihood phylogenies: assessing the perfor- in America North of Mexico. Smithsonian Institute Press, mance of PhyML 3.0. Syst Biol 59:307–321 Washington 123 Invasive winter moth in the northeastern United States 2889

Levine JM, Adler PB, Yelenik SG (2004) A meta-analysis of Roland J (1988) Decline in winter moth populations in North biotic resistance to exotic plant invasions. Ecol Lett America—direct versus indirect effect of introduced par- 7:975–989 asites. J Anim Ecol 57:523–531 Lewis MA, Kareiva P (1993) Allee dynamics and the spread of Roland J (1990) Interaction of parasitism and predation in the invading organisms. Theor Popul Biol 43:141–158 decline of winter moth in Canada. In: Watt A, Leather SR, Macphee A, Newton A, McRae KB (1988) Population studies Hunter AF (eds) Population dynamics of forest insects. on the winter moth Operophtera brumata (L.) (Lepi- Intercept Ltd, Andover, pp 289–301 doptera, Geometridae) in apple orchards in Nova Scotia. Roland J, Embree DG (1995) Biological control of the winter Can Entomol 120:73–83 moth. Annu Rev Entomol 40:475–492 Maron JL, Vila` M (2001) When do herbivores affect plant Royama T (1981) Evaluation of mortality factors in insect life invasion? Evidence for the natural enemies and biotic table analysis. Ecol Monogr 51:495–505 resistance hypotheses. Oikos 95:361–373 RstudioTeam (2015) RStudio: integrated development envi- Maron JL, Harrison S, Greaves M (2001) Origin of an insect ronment for R. RStudio, Inc., Boston outbreak: escape in space or time from natural enemies? Sakai AK, Allendorf FW, Holt JS, Lodge DM, Molofsky J, With Oecologia 126:595–602 KA, Baughman S, Cabin RJ, Cohen JE, Ellstrand NC, Matosevic D, Melika G (2013) Recruitment of native parasitoids McCauley DE, O’Neil P, Parker IM, Thompson JN, Weller to a new invasive host: first results of Diyocosmus kur- SG (2001) The population biology of invasive species. iphilus parasitoid assemblage in Croatia. Bull Insectol Annu Rev Ecol Syst 32:305–332 66:231–238 Schaefer PW, Fuester RW, Chianese RJ, Rhoads LD, Tichenor Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES RB (1989) Introduction and North-Amherican establish- Science Gateway for inference of large phylogenetic trees. ment of Coccygomimus disparis (Hymenoptera, Ichneu- In: Proceedings of the Gateway computing environments monidae), a polyphagus pupal parasite of Lepidoptera, workshop (GCE). New Orleans including gypsy moth. Environ Entomol 18:1117–1125 Morin H (2015) Winter moth (Operophtera brumata L.) natural Schenk D, Bacher S (2002) Functional response of a generalist enemy diversity and abundance in infested areas in Mid- insect predator to one of its prey species in the field. J Anim coast Maine. Thesis in Biology, The University of Maine Ecol 71:524–531 Murdoch WW (1969) Switching in general predators: experi- Schonrogge K, Stone GN, Crawley MJ (1995) Spatial and ments on predator specificity and stability of prey popu- temporal variation in guild structure—parasitoids and lations. Ecol Monogr 39:335 inquilines of Andricus quercuscalicis (Hymenoptera, Murdoch WW, Chesson J, Chesson PL (1985) Biological-con- Cynipidae) in its native and alien ranges. Oikos 72:51–60 trol in theory and practice. Am Nat 125:344–366 Shea K, Chesson P (2002) Community ecology theory as a Myers JH (2018) Population cycles: generalities, exceptions and framework for biological invasions. Trends Ecol Evol remaining mysteries. P R Soc B 285:20172841 17:170–176 Myers JH, Cory JS (2013) Population cycles in forest lepi- Silvestri F (1941) Contribuzioni alla conoscenza degli insetti doptera revisited. Ann Rev Ecol Evol Syst 44:565–592. dannosi e dei loro simbionti. VI. La falena brumale o la https://doi.org/10.1146/annurev-ecolsys-110512-135858 brumale (Operophtera brumata L.). Bolle Lab Entomol Nechols JR, Kauffman WC, Schaefer PW (1992) Significance of Portici 5:61–120 host-specificity in classical biological control. In: Kauff- Starrett J, Hedin M (2007) Multilocus genealogies reveal mul- man WC, Nechols JR (eds) Selection criteria and ecolog- tiple cryptic species and biogeographical complexity in the ical consequences of importing natural enemies. California turret spider Antrodiaetus riversi (Mygalomor- Entomological Society of America, Lanham, pp 41–52 phae, Antrodiaetidae). Mol Ecol 16:583–604 Oaten A, Murdoch WW (1975) Switching, functional response, Strauss A, White A, Boots M (2012) Invading with biological and stability in predator–prey systems. Am Nat weapons: the importance of disease-mediated invasions. 109:299–318 Funct Ecol 26:1249–1261 Pearsall IA, Walde SJ (1994) Parasitism and predation as agents Sullivan DJ (1987) Insect hyperparasitism. Annu Rev Entomol of mortality of winter moth populations in neglected apple 32:49–70 orchards in Nova Scotia. Ecol Entomol 19:190–198 Swofford DL (2003) PAUP*. Phylogenetic analysis using par- Pearson DE, Ortega YK, Eren O, Hierro JL (2018) Community simony (*and other methods). Version 4. Sinauer Associ- assembly theory as a framework for biological invasions. ates, Sunderland Trends Ecol Evol 33:313–325 Tenow O, Nilssen AC, Bylund H, Pettersson R, Battisti A, Bohn Poulin R (2017) Invasion ecology meets parasitology: advances U, Caroulle F, Ciornei C, Csoka G, Delb H, De Prins W, and challenges. Int J Parasitol Parasites Wildl 6:361–363 Glavendekic M, Gninenko YI, Hrasovec B, Matosevic D, Rambaut A (2014) FigTree Tree Figure Drawing Tool http:// Meshkova V, Moraal L, Netoiu C, Pajares J, Rubtsov V, tree.bio.ed.ac.uk/software/figtree/. Accessed 2 May 2018 Tomescu R, Utkina I (2013) Geometrid outbreak waves RCoreTeam (2013) R: a language and environment for statis- travel across Europe. J Anim Ecol 82:84–95 tical computing. R Foundation for Statistical Computing, Tobin PC, Berec L, Liebhold AM (2011) Exploiting Allee Vienna effects for managing biological invasions. Ecol Lett Richardson DM, Pysek P (2006) Plant invasions: merging the 14:615–624 concepts of species invasiveness and community invasi- Townes HK (1940) A revision of the Pimplini of eastern North bility. Prog Phys Geogr 30:409–431 America (Hymenoptera: Ichneumonidae). Ann Entomol Soc Am 33:283–323 123 2890 H. J. Broadley et al.

Townes H (1969) The genera of Ichneumonidae, part 1. Whited BM (2007) The population ecology of winter moth Mem.Am Entomol Inst 11:i–ii ?1–300 (Operophtera brumata) in eastern Massachusetts. Masters Townes HK, Townes M, Walley GS, Townes G (1960) Ich- thesis in Organismic and Evolutionary Biology, University neumon-flies of America north of Mexico: 2 Subfamily of Massachusetts, Amherst Ephialtinae, Xoridinaem Acaenitinae. US Natl Mus Bull Wickham H (2009) ggplot2: elegant graphics for data analysis. 216:1–676 Springer, New York Van Driesche RG (1983) Meaning of ‘‘percent parasitism’’ in Wylie HG (1960) Insect parasites of the winter moth, Oper- studies of insect parasitoids. Environ Entomol ophtera brumata (L.) (Lepidoptera: Geometridae) in 12:1611–1622 western Europe. Entomophaga 5:111–129 Varley GC, Gradwell GR, Hassell MP (1973) Insect population Yu DSK, van Achterberg C, Horstmann K (2012) World Ich- ecology: an analytical approach. Blackwell Scientific, neumonoidea 2011. Taxonomy, biology, morphology and Oxford, England, 212 pp distribution. Taxapad. Home of , Nepean, Vindstad OPL, Hagen SB, Jepsen JU, Kapari L, Schott T, Ims Ontario RA (2011) Phenological diversity in the interactions Zappala L, Bernardo U, Biondi A, Cocco A, Deliperi S, Delrio between winter moth (Operophtera brumata) larvae and G, Giorgini M, Pedata P, Rapisarda C, Garzia GT, Siscaro parasitoid wasps in sub-arctic mountain birch forest. Bull G (2012) Recruitment of native parasitoids by the exotic Entomol Res 101:705–714 pest Tuta absoluta in Southern Italy. Bull Insectol Vindstad OPL, Schott T, Hagen SB, Jepsen JU, Kapari L, Ims 65:51–61 RA (2013) How rapidly do invasive birch forest geometrids recruit larval parasitoids? Insights from comparison with a Publisher’s Note Springer Nature remains neutral with sympatric native geometrid. Biol Invasions 15:1573–1589 regard to jurisdictional claims in published maps and Waage J, Greathead D (1985) Insect parasitoids. Academic institutional affiliations. Press Limited, San Diego

123